This article was downloaded by: [Wageningen UR Library]On: 12 January 2015, At: 12:14Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

The Journal of Agricultural Educationand ExtensionPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/raee20

Learning and Innovation in Agricultureand Rural Development: The Use ofthe Concepts of Boundary Work andBoundary ObjectsTalis Tisenkopfsa, Ilona Kundaa, Sandra šūmanea, Gianluca

Brunorib, Laurens Klerkxc & Heidrun Moschitzd

a Baltic Studies Centre, University of Latvia, Kokneses prospekts26-2, Riga, LV 1014, Latviab Department of Agriculture, Food and Environment (DAFE),University of Pisa, Via del Borghetto 80, Pisa 56124, Italyc Knowledge, Technology and Innovation Group, WageningenUniversity, PO Box 8130, Wageningen, EW 6700, The Netherlandsd Department of Socio-Economics, Research Institute of OrganicAgriculture FiBL, Ackerstrasse 113, PO Box 219, Frick 5070,SwitzerlandPublished online: 08 Jan 2015.

To cite this article: Talis Tisenkopfs, Ilona Kunda, Sandra šūmane, Gianluca Brunori, Laurens Klerkx& Heidrun Moschitz (2015) Learning and Innovation in Agriculture and Rural Development: The Useof the Concepts of Boundary Work and Boundary Objects, The Journal of Agricultural Education andExtension, 21:1, 13-33, DOI: 10.1080/1389224X.2014.991115

To link to this article: http://dx.doi.org/10.1080/1389224X.2014.991115

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,

proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Learning and Innovation in Agricultureand Rural Development: The Use ofthe Concepts of Boundary Work andBoundary Objects

TALIS TISENKOPFS*, ILONA KUNDA*, SANDRA ŠŪMANE*,GIANLUCA BRUNORI†, LAURENS KLERKX‡ and HEIDRUN MOSCHITZ§

*Baltic Studies Centre, University of Latvia, Kokneses prospekts 26-2, Riga, LV 1014, Latvia, †Department ofAgriculture, Food and Environment (DAFE), University of Pisa, Via del Borghetto 80, Pisa 56124, Italy,‡Knowledge, Technology and Innovation Group, Wageningen University, PO Box 8130, Wageningen, EW6700, The Netherlands, §Department of Socio-Economics, Research Institute of Organic Agriculture FiBL,Ackerstrasse 113, PO Box 219, Frick 5070, Switzerland

ABSTRACT Purpose: The paper explores the role of boundary work and boundary objectsin enhancing learning and innovation processes in hybrid multi-actor networks forsustainable agriculture (LINSA).Design/Methodology/Approach: Boundary work in LINSA is analysed on the basis of sixcase studies carried out in SOLINSA project under a common methodology. Indeveloping typologies of boundary work and objects, a grounded approach is used.Findings: LINSA analysis demonstrates the dynamic character, diverse forms and multiplefunctions of boundary work and objects in three domains: learning, innovation, andsustainability. Addressing specific types of goals and actors leads to specific types of boundarywork and boundary objects. Context-appropriate boundary work allows aligning differingactor attitudes, gaining increased external support, and developing LINSA. The concepts ofboundary work and boundary objects are relevant in a broad range of divergent LINSAsettings. Boundary work has its limitations, but its facilitation supports reaching LINSA goals.Practical Implications: The paper proposes recognising context-appropriate forms ofboundary work and skilful use of emerging boundary objects to both promote internalconsolidation of LINSA and effective external communication to foster learning andinnovation for sustainability.Originality/Value: The paper provides insights into the forms, dynamic and outcomes ofboundary work in LINSA in three key domains: developing shared knowledge base, co-producing innovation and negotiating sustainability.

KEY WORDS: LINSA, Boundary work, Boundary objects, Learning, Innovation, Sustainability

Corresponding address: Talis Tisenkopfs, Baltic Studies Centre, University of Latvia, Kokneses prospekts 26-2,Riga, LV 1014, Latvia. Tel: +371-29417173, Fax: + 371-67470244. Email: talis.tisenkopfs@lu.lvThe views presented here are the views of the authors and do not necessarily represent the views of the Commission.

Journal of Agricultural Education and ExtensionVol. 21, No. 1, 13–33, February 2015

1389-224X Print/1750-8622 Online/15/010013-21 © 2014 Wageningen Universityhttp://dx.doi.org/10.1080/1389224X.2014.991115

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

1. Introduction

Innovation processes are increasingly conceptualized as the outcome of collaborativenetworks where information is exchanged and learning processes happen (Knickel et al.2009, 139). The SOLINSA project explored learning and innovation in specific kinds ofnetworks – LINSA, defined as Learning and Innovation Networks for SustainableAgriculture, which involve actors who are mutually engaged to achieve common goalsfor sustainable agriculture and rural development by cooperating, sharing resources andco-producing new knowledge (Brunori et al. 2013). LINSA represent learning andinnovation in heterogeneous multi-actor environments as they consist of actors fromdifferent agro-food domains: producers, consumers, experts, non-governmental organiza-tions, enterprises, local administrations and components of the formal agriculturalknowledge system. Taking into account such diversity of actors and their dispositions,multi-actor interactions and co-construction of new shared meanings are centralcomponents in learning and innovation in LINSA.

In order to better understand and facilitate agricultural learning and innovation inmulti-actor networks we need to examine mechanisms that allow their joint endeavours,connections between diverse life-worlds and knowledge bases. The idea increasinglyapplied to examine such processes is that of boundary work and its components,boundary objects among these. Boundary work helps to manage all kind of demarcationsbetween and within different actors and social groups and align their differentmotivations, perceptions and expectations into common cognitive frames and concertedactions (Clark et al. 2011; Klerkx, Van Mierlo, and Leeuwis 2012). The paper addressesthe following research question: what is the role of boundary work and boundary objectsin enhancing learning and innovation in multi-actor networks for sustainable agriculture.

We go beyond the original tradition of science and technology studies which viewboundary work as an element of communication and knowledge transfer between scienceand non-science (Gieryn 1983; Moore 1996; Guston 2001) and look at interactions withina more diverse set of actors, notably at interactions among practitioners themselves, peer-to-peer relations, involvement of civic groups, local administrations, etc. We use theconcept of boundary work to highlight exchange and translation of individual knowledgeinto collectively shared, durable knowledge and innovations for sustainable agriculture,which go hand in hand with the development of networks themselves.

In particular we analyse the processes and outcomes of boundary work in connectingand aligning LINSA actors across different social, institutional and conceptual boundariesin the three key domains of LINSA: joint learning (Carr and Wilkinson 2005; Hornidge,Ul-Hassan, and Mollinga 2009; Oreszczyn, Lane, and Carr 2010), innovation (Klerkx andLeeuwis 2009; Leeuwis and Aarts 2011) and negotiating agricultural sustainability (Cashet al. 2003; Carolan 2006; Mollinga 2010; Clark et al. 2011). These three domains areclosely interrelated but each of them also illuminates a different aspect and anintervention area of boundary work. Boundary work in learning is primarily aboutconstruction of common cognitive frames: alignment of ideas, perceptions, interpreta-tions, meanings. Boundary work in innovation is more focused on joint action formationor alignment in practices. Boundary work regarding sustainability demonstrates bothcognitive and practical alignment in such a highly contested arena as sustainability is, andwhich may demand particular efforts from multi-stakeholder networks to develop andimplement shared holistic solutions. In addition, in all three domains we analyse

14 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

boundary work and objects in a dynamic perspective and capture their changing purpose,form, scope and outcome along the life-span of networks.

The article has four sections. The first section provides theoretical input toconceptualize boundary work and boundary objects. The second section characterizescases and outlines the methods used to analyse boundary work. The third section exploresthe manifestations of boundary work and boundary objects to promote processes andoutcomes on learning, innovation and sustainability in LINSA. The fourth section reflectson the theoretical and practical implications of our findings with the view of helpingLINSA to perform boundary work more efficiently.

2. Conceptual Framework to Study Boundary Work and BoundaryObjects in LINSA

Following recent concepts of innovation as a systemic activity (e.g. Knickel et al. 2009;Klerkx, van Mierlo, and Leeuwis 2012) SOLINSA research addressed agriculturallearning and innovation from a network approach which allows acknowledgement andintegration of various knowledge actors, sources, types, and processes and learningmodes. Thus, we shift from monoculture of scientific knowledge towards ecology ofknowledge, which assumes the diversity of knowledge and its composite character(Santos, Nunes, and Meneses 2007; Bruckmeier and Tovey 2008). It allows consideringfarmers, but also consumers, rural residents, market enterprises, NGOs, policy makersand other actors with their diverse knowledge inputs to agricultural development (Knickelet al. 2009; Oreszczyn, Lane, and Carr 2010; Leeuwis and Aarts 2011; Renting et al.2008). The network model advocates active social learning (Wenger 2000; Leeuwis andAarts 2011), which stresses that knowledge is contextually grounded and acquired ininteraction and negotiation between actors. To achieve learning and innovation in hybridnetworks, actors have to align their diverse attitudes, motivations and values into a sharedknowledge pool and collective or concerted action. This is of particular relevanceregarding the highly contested concept and practices of sustainable agriculture(Koutsouris 2008; Hermans et al. 2010).

It has become acknowledged that building and managing learning and innovation inmulti-stakeholder networks requires boundary work that would improve connectionsof different life-worlds, facilitate learning across their boundaries and transformation ofknowledge into innovation (Klerkx and Leeuwis 2009; Clark et al. 2011; Klerkx, VanMierlo, and Leeuwis 2012; Mollinga 2010). Boundary work is needed because tensionsarise at the interface between actors with different views of what constitutes reliable oruseful knowledge (Clark et al. 2011). Diversity and conflicts present in such networks canbe enriching and facilitate creativity and innovation, but they can also block up new ideasand initiatives if common interest and language among actors cannot be established(Ring, Peredo, and Chrisman 2010). Communication failures are frequent whendifferences in objectives, interests, languages, frames are wide and no interfaces betweenthem are found. This calls for management or enabling of knowledge processes acrossboundaries, i.e., boundary work.

In exploring boundary work, there are some shared hypotheses on what promotes itssuccess, namely participation of stakeholders in agenda setting and knowledge produc-tion, accountability of governance, and the use of boundary objects (Clark et al. 2011).However there is a shortage of analysis of successful strategies in particular settings

Learning and Innovation in Agriculture and Rural Development 15

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

(Clark et al. 2011). Our study contributes to exploring boundary work and its outcomes,focussing on ‘multiple heterogeneous communities of knowledge producers’ (Clark et al.2011, 2) which are inside or outside LINSA - self-governing entities in sustainableagriculture, with a considerable component of learning and innovation in their activities.It is a distinctive setting, which has been explored very little so far, thus our study aims toidentify some distinctive traits in boundary work strategies and outcomes.

We target the participation processes and boundary objects used in boundary work, andaim to view the kinds of interactions used to foster connections between agents,constellations of agents that participate, and outcomes achieved - for the three LINSAactivity domains (joint learning, innovation, sustainability) and the network on the whole.

To characterize the boundary work interactions, we use as the starting point ideas fromWenger (2000, 236–237), who distinguishes between four types:

. encounters - visits, meetings, and in short anything providing exposure to a differentpractice;

. boundary practice - a special practice aimed at maintaining the crossing ofboundaries;

. peripheries – practices to allow connecting for interested outsiders or potentialmembers;

. specific tasks/projects - kinds of practice that result from addressing non-routinetopics, bringing together novel kinds of participants and transforming roles.

We aim to see the peculiarities of these interactions in the multi-stakeholder, evolvingcontext of LINSA, thus linking it also to the issue of participants, i.e. uncovering typicalconstellations of agents.

Certainly the interactions are aimed to produce specific outcomes, which for LINSAare related to learning, innovation, negotiation of sustainability and the networkdevelopment as such. We will expand on these when discussing boundary objects, athird important component of boundary work, which brings together boundary workinteractions and their participants.

A boundary object is defined as ‘an entity shared by several different communities butviewed or used differently by each of them, being both plastic enough to adapt to localneeds and the constraints of the several parties employing them, yet robust enough tomaintain a common identity across sites’ (Star and Griesemer 1989, 393). These aretangible or intangible - a trademark, a publication, a code of practice, a website, astrategic paper, an idea etc. – and are of shared interest for participants and thereforecreate interfaces for their communication, interaction and coherence (Star and Griese-mer 1989).

Boundary objects cannot be imposed; they emerge through the process of interaction(Star and Griesemer 1989, 393); the process of joint development of a boundary objectfosters interaction through verbal and non-verbal communication.

A very general categorization of boundary objects (Wenger 2000) is that of artefacts,discourses and processes. For our purposes, it is important to identify typicalmanifestations of various categories of boundary objects in LINSA, as well as theirevolution.

With regard to LINSA activity domains, it is important that the key feature of boundaryobjects is their interpretive flexibility which allows actors to find their own meaning inthem, thus furthering learning through discussion with other actors of the different

16 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

interpretations (Klerkx, van Mierlo, and Leeuwis 2012). For instance, boundary objectscan help to align different frames in issue-specific learning (Tisenkopfs, Kunda, andŠūmane 2014).

Boundary objects also act as vehicles for change by enabling networks of actors toalign around a certain vision they espouse, negotiate a shared direction, and enhancecollaboration in innovation processes (Klerkx et al. 2012; Theodorakopoulos et al. 2014).Similarly Guston (2001, 400) notes that boundary objects may change practices onboth sides of a boundary, although this requires consent of agents and perhaps broadercultural changes.

Movement toward change may be made more evident by distinguishing betweengeneral and specific boundary objects. A general boundary object is about ‘ground rules’and is created in early stages of managing of a boundary, while specific boundary objectsare created to deal with more fine-tuned, context-specific issues. There is a certainevolution of boundary objects and boundary work, which also links boundary work tonetwork development (from early to more evolved stages, as more complex boundarywork is carried out and achieves goals).

Thus boundary work is related to certain outcomes for interaction participants and thewhole network. We will explore these outcomes in learning, innovation, negotiation ofsustainability and network development in LINSA.

3. Data and Methods of Analysis

A longitudinal collaborative research with LINSA was conducted over three years toobserve and analyse the evolution of network learning, innovation and negotiatedmeaning of sustainability. The data was gathered in a two-stage process. Stage 1concentrated on exploration of LINSA properties: scope, origin, function, networkcomposition, learning structures employing participatory research methods (stakeholderworkshops, study trips with practitioners, participant observation, feedback on researchfindings, interviews). Stage 2 focussed explicitly on a number of boundary work andobject examples from each LINSA. Specific descriptions of boundary work and boundaryobjects were produced for 17 LINSA describing the kinds of boundaries crossed, the keyactors involved, forms and outcomes of boundary work. Evolution of boundary objectsand boundary work was observed and examined.

For the purposes of this article we chose six cases, which provide the richest empiricalevidence of boundary work. The following selection criteria were used: (1) the intensityand types of the boundary; (2) the diversity of actors involved; (3) the variety ofboundary objects; and (4) outcomes of boundary work in terms of contribution tolearning, innovation and sustainability. The selected cases represent agricultural networksof varied origin, size, stakeholder diversity, objectives of innovation, and approaches tosustainability from agriculturally focussed to holistic ones. They are related to differentareas of agriculture, food production and consumption, rural development and involvenew urban-rural linkages. The analysed networks represent intensive learning processesand cross-linkages between different domains of knowledge. The innovations are often acombination of technical, economic, organizational, social and political innovation andsustainability concerns are explicit in all networks, however in some LINSA not clearlyagreed. A broad range of learning and innovation needs, structural differences in the

Learning and Innovation in Agriculture and Rural Development 17

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

networks, various knowledge bases and interpretations of action call for negotiation ofmeanings, reconciliation of interests and alignment of practices.

The cases analysed were grouped according to the main focus of boundary work –learning, innovation, sustainability – and each dimension was represented by two cases inorder to characterize variations. Importantly, boundary work and objects have multipleand combined impacts on learning, innovation and sustainability processes and outcomes;however, for analytical purposes, we illustrate their contributions to a single most notabledomain in a given LINSA. In the Table 1 we provide an overview of boundary workmanifestations in the examined LINSA which were systematically compared andreiteratively examined by groups of researchers and LINSA participants.

4. Using Boundary Work and Objects in LINSA to Enhance Learning,Innovation and Sustainability

In this section we analyse how boundary work takes place in networks in specificdomains of learning, innovation and sustainability, what are the typical boundaryinteractions, participant constellations and boundary objects used.

4.1. Boundary Work for Learning

For active and relevant learning to occur, the network participants have to negotiate anumber of boundaries, between various (1) stakeholder groups representing variousknowledge bases (e.g. farmers, researchers, advisors); (2) learning issues (technical,economic, organizational, political) as they may change during the course of innovation;(3) learning forms (e.g. face-to-face and technically mediated); and (4) attitudes towardslearning outcomes, mistakes and difficulties. This means that boundary work has to becarried out to maximize the resources present in the network and stimulate the learningnecessary to achieve the network’s goals. Boundary work would include variousinteractions to align network participants’ needs, expectations, perceptions and practices,to allow new knowledge to be developed. Two cases demonstrate how boundary workand boundary objects enhance learning in complex agricultural networks.

The Latvian Fruit growing network consists of diverse actors ranging from orchardowners, farmers’ associations, cooperatives, research institutes, public organizations,retailers and consumers engaged with ideas and practices of integrated fruit-growing. Thenetwork is aimed at development of fruit growing sector in Latvia and promotion of themarket of local fruit. It is geographically spread across the country and has severalterritorial sub-networks of more intensive learning and innovation.

Learning in the LINSA unfolds in several interrelated directions: fruit-growingtechnologies, economic organization of supply chain, use of state support, education ofconsumers. In the recent years with the development of new orchards and theestablishment of producer groups, learning intensifies into the economic domain andrelates to organization of collective marketing, sales, branding, developing the links withconsumers.

Boundary interactions around these issues are intense and involve various stakeholdersin different combinations. Learning about agricultural policies happens in interactionbetween farmers’ organizations and policy makers; new methods in horticulture are learntin mutual exchange among farmers and between the farmers and researchers; knowledge

18 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Table 1. Boundary work manifestations in LINSA depending on boundary work focus

BWfocus

BWManifestationsLINSA andcountry

Key actors inboundary work:Farmers ConsumersAdvisorsResearchers Civicgroups Professionalassociations SMEsPolicy makersLocal authorities

Main forms of boundarywork: Encounters PracticesPeripheries Specific work

Main outcomes ofboundary work

Learning

Fruit growingnetwork, Latvia

FarmersResearchersProfessionalassociations

Encounters (field days,seminars, visits)Peripheries (Apple Festival,Cherry Days)Specific work (research anddemonstration projects)

Improvedpractices inintegrated fruit-growing

Charter of GoodPractices,France

FarmersAdvisorsResearchersProfessionalassociations

Encounters (informalinterviews, meetings)Boundary practices (jointconsultations at local regionaland national level, reviewingof the Charter)

Development andadoption ofsustainablepractices in cattleproduction

Inno

vatio

n

Cooperative ofhealth carefarms ‘Boer enZorg’, TheNetherlands

FarmersPolicy makersLocal authoritiesProfessionalassociationsResearchersAdvisors

Encounters (meetings, visits,seminars)Boundary practices(cooperative maintaininglinks with local authorities)Specific work (commissionedresearch to confirm value)

Development ofcare farming as aservice betweentwo institutionalsystems

AssociationCrisoperla, Italy

Farmers,ConsumersAdvisorsLocal authoritiesProfessionalassociationsCivic groupsSMEsResearchers

Encounters (meetings, visits,seminars)Peripheries (workshops,drafting of policy proposals,fairs, seeking links with newcivic groups)Boundary practices(participation in regional andnational associations)Specific work (developing acode of practice)

Redefinedmeaning andpractices oforganic farming,involving notonly technical,but social, ethicaland politicalaspects.

Sustainability

Brighton andHove FoodPartnership, UK

Civic groupsConsumersSMEsLocal authorities

Encounters (partnershipdevelopment meetings)Peripheries (newsletter,publicity activities)Specific work (groupprojects, college course)

Redefinedunderstanding ofsustainable well-being in the city;good practicesconveyed widely

Biogas network,Latvia

ResearchersFarmersSMEs

Encounters (field days,informal meetings)Specific work (commissionedresearch, demonstrationprojects)

Continueddisagreement ondifferent aspectsof sustainability(environmental,social, economic)

Learning and Innovation in Agriculture and Rural Development 19

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

about processing is acquired from food technologists and experienced producers;marketing knowledge is generated in farmers’ interaction with retailers and cooperatives.

The Fruit LINSA demonstrates the presence of three kinds of boundary work:encounters, peripheries and specific work. Encounters (field days, study tours,demonstrations, seminars, individual consultations) are a particularly effective form ofboundary work as they bring together farmers and researchers in direct face to facelearning situation. This form was found most widely used in LINSA for technicallearning issues (growing technologies, varieties, plant protection, etc.). This form ofboundary work is highly praised by the farmers as it provides hands on experientiallearning and is also emotional. Projects implemented by research institutes (appliedresearch and demonstration projects) and cooperatives (investment projects) aremanifestations of special boundary work which allows to undertake systematic learningactivities (educational seminars, training courses, study visits abroad, farmers commis-sioned researches etc.) over a longer period of time. Many boundary work activities infruit network are technically mediated (e.g. information retrieval on websites of researchinstitutes and farmers’ organizations, web based training courses, remote serviceprovision for farmers using internet) and become available to farmers due to expansionof internet based knowledge services and enhancement of farmers computer skills.

Integrated fruit growing is a general boundary object whose various aspects – growingtechnology, processing methods, marketing infrastructure, supportive public policy,consumers demand – are explored, discussed and redefined in the network’s interactions.Different more specific boundary objects are generated in sequence corresponding tothe learning needs of individual members and to the overall evolution of the network.In the initial phase of LINSA development when the orchards were planted, appropriateapple varieties and cultivation techniques were specific boundary objects discussedamong farmers and researchers. When the production became more established themarketing issues became topical and the network members interacted around boundaryobjects such as marketing infrastructure, storage facility, cooperative sales, contracts withretailers, etc. The Fruit network is peculiar as it develops and actively uses culturallyattuned boundary objects – festivals and public events (e.g. Apple Festival, Cherry Days,Scientists night) to connect with the wider public and promote consumer education aboutfruit sector and sustainable food production and consumption issues. In a hybrid network,specific boundary objects may be activated in specific bilateral or multilateral interaction,e.g. quality standards and kinds of apple varieties are a notable boundary object innegotiations with retailers; while crown shaping techniques are negotiated in consulta-tions and field days between researchers/consultants and farmers.

Thus, boundary interaction enhances both learning processes and role changes –researchers become consultants, farmers co-innovate with researchers, some consumersstart gardening, children become educated about sustainable diets through School Fruitprogramme, etc.

Inclusive boundary work and objects enhance participation in the network and makesit more egalitarian, especially enhancing learning in the network peripheries (Laveand Wenger 1991). However, boundary work does not happen by itself, the leadingpersonalities or organizations (boundary brokers) are needed. In the Fruit LINSA thedriving actors for boundary work are fruit researchers committed to knowledgesharing, the leaders of producers’ groups, and individual farmers who are open to sharetheir experiences with peers. This ‘triangulation’ between the researchers, farmers’

20 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

organizations and individual farmers open for knowledge exchange makes boundarywork effective and helps to bridge the traditional research-practice gap.

The Charter of Good Practices is a French network of farmers, farmers’ organizations,research and extension organizations built as a collective answer to the mad cow diseasecrisis. The idea was to foster the quality and the security of cattle farmers’ practices andto communicate these to the beef and dairy industry, the retailers, and the general public.The task was two-fold: to formulate the principles of improved cattle-farming, and get acritical mass of farmers to implement them.

The initiative was started by farmer organizations. They established a coalition withinstitutions of formal agricultural knowledge system, which took a leading role innetwork building. Development of the Charter was a multi-level and multi-actor processbringing together all the stakeholders related to milk and beef production andconsumption: the dairy and beef industry, retailers, consumer and citizen associations,regulation makers. To reconcile all the interests and concerns, a complex boundary workwas carried out in the form of stakeholder meetings, development and implementation offarmers’ and their advisors’ assessment system. It was accompanied by technical work,expert consultations and, finally, synthesis, negotiation and decision-making regardingthe practice-generated issues.

Most of the boundary work was made through the involvement of the advisers whopresented the Charter to the farmers and proposed them to evaluate their practices and toadhere to the Charter. A state funding of the facilitation and advisory activities positivelyinfluenced the farmers learning and adoption of new practices. Nowadays the networkinvolves about 110,000 farmers and 3,000 facilitators and about 90% of the dairy farmersand 66% of the beef farmers in France belong to the Charter.

The main boundary object in this case is the Charter itself: a 12 pages leaflet,describing the 41 good practices the farmers commit to follow. Both for network internaland external use the Charter is made available on website and leaflets and logos are usedas boundary objects to communicate with external actors. The Charter is periodicallyrenewed in consultations among the stakeholders and improved descriptions of practicesare made available to farmers through website and advisory work. Thus continuousboundary work results in updating of boundary objects. However, not all farmers,especially small and isolated ones, are involved in these boundary interactions due to thefact that they are not clients of the advisory organization.

This case demonstrates the role of research and extension organizations in boundarywork and rather top down manner of its organization. Partly this explains the scale of thenetwork. Extensive boundary work was necessary to negotiate the Charter in thebeginning and introduce novelties at later stages. The boundary work enhanced learningamong farmers and changes in their practices and resulted in gradual changes within thewhole cattle farming system. The example demonstrates how well coordinated boundarywork activates learning, adoption of new practices and changing a regime.

Both cases suggest that the idea and practices of boundary work and objects illustratethe dynamic of learning – from unfamiliarity and lack of negotiated joint meaning toincreased commonality through intensive negotiation of meaning (the object ofnegotiation being a boundary object at this point) to gradually arriving at a similar levelof familiarity, agreed definition of what the object means and how to deal with it (at thispoint the boundary object becomes part of a shared practice). Another way to look atboundary work is to view it as a process of jointly filling up an empty knowledge space

Learning and Innovation in Agriculture and Rural Development 21

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

with jointly created knowledge, e.g. a new practice, a method being a boundary object. Tosum up, we argue that boundary work co-evolves with the learning priorities anddevelopment of the network; boundary objects change with the flow of the learning andinnovation and in turn influence it. There might be a hierarchy and/or sequence inboundary objects produced. The role of leading boundary work organizations and personswas notable in both cases as well as purposeful use of boundary work and objects in peer-to-peer communication and external communication of the networks to introducesustainable practices to other food chain actors, notably consumers.

4.2. Boundary Work for Innovation

Boundary work is instrumental for innovation development in LINSA and the creation ofits support network. Establishing connections through boundary work can help evenradical innovations spread actively and continue developing in new contexts, giving riseto satellite innovations. For that purpose the network needs mobilising new supporters,balancing the interest in economic gains and other values (social, ethical, knowledgedevelopment, etc.), and skills to negotiate expanded innovation goals. Thus boundarywork may be needed with regard to (1) involving new stakeholders in the network toaccept innovation, (2) gaining new supporters or users of the innovation outside of thenetwork, as well as (3) adjusting the network goals to include new innovation goals, oncethe initial goal has been achieved, and/or new opportunities are identified. SOLINSAresearch identified the relevant boundary work processes and objects used to negotiateand promote innovation, and their impacts on LINSA. We illustrate it through twodiffering LINSA cases.

The case of the Dutch cooperative of health care farms ‘Boer en Zorg’ demonstratesboundary work to promote a radical innovation both internally (within the LINSA) andexternally (with public servants), the process in its turn influencing the development ofthe network itself. The innovation in question resides at the boundary of two knowledgeworlds: agriculture and health. Care farmers use agricultural resources, such as landscape,animals, gardens, and forests to address the needs of people with health problems,reintegrate them into social and economic life and improve their wellbeing. Thus thedevelopment of innovation requires boundary work both to ensure its legitimacy andacceptance with policy actors and other relevant external stakeholders, and in parallel ithas to gain new participants who implement the innovation and maintain a sharedknowledge base among the existing ones.

In addressing the issues with external policy actors (on national and increasingly onlocal level) the LINSA makes use predominantly of encounters and special persuasionactivities. The latter involve using and commissioning research to demonstrate thebenefits of the innovation, to negotiate funding arrangements, and to lobby. Special workoften takes the form of externally funded projects. Participants include LINSA activistsand managers, researchers, and policy actors. The boundary objects are predominantlydocuments, but also concepts and ideas around the notion of care farming. With theprospect of more decentralization in health care and increasing the role of local levelgovernment, the boundary work has shifted from general boundary object (what is thebenefit of care farming) to very specific boundary object representing negotiated locallevel administrative and funding arrangements.

22 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Internally, boundary work is required to ensure a shared standard of implementing theinnovation, as care farming has its own certification. The LINSA includes several typesof care farmers, differing in their backgrounds (agriculture, health care, other) and interestin developing the practice. Thus concerted boundary work is needed to negotiate thesediverse knowledge bases and goals. The participants of boundary work are mostly LINSAmembers, with some external experts delivering specific information at training sessions.Another subset of participants is the accrediting institutions, which issue certificates tocare farmers. The boundary work types are both encounters, practice-based (demonstra-tions, study visits) and special work (training sessions and courses). The outcome of theboundary work is prescriptive, as care farmers have to subscribe to certain standards ofpractice. The boundary objects in this strand of activities range from the concept of carefarming and its components, acceptable practices, and most of these are documented intangible form; there is also administrative software to be used by all. The evolution maybe seen in further developing various standards of care farming activity to makeimplementation of the innovation both easier and of consistent quality.

With regard to impact on the network, the above development of the innovationhas happened alongside with a search for appropriate governance form. The LINSA hasgrown in terms of members, clients and geographical coverage. Simultaneously, the rapidgrowth has put pressures on its cohesiveness which shows the need to balance externaland internal boundary work and specifically the importance of internal one to establishshared understandings among LINSA members, facilitate their loyalty and engagement.Overall, the outcome of the external boundary work was successful institutionalization ofan innovative service at the boundary of two knowledge systems – agriculture and healthcare. It has contributed to the development of LINSA both in scope, shared meanings,and improved practices. Development of the LINSA has also meant dealing with newchallenges, again requiring boundary work.

The case of the Italian Association Crisoperla is focussed on promoting a combinationof economic and social innovation – organic farming and solidary economy in a specificregion with previously scarce social interaction among farmers and citizens/consumers.The boundary work is targeting both external stakeholders – policy makers and thegeneral public, and internal – farmers, fishermen, direct purchasing groups, citizen groupsconcerned with sustainable lifestyle issues. The distinctive trait of Crisoperla case isthe involvement of the civic component, the ethical dimension of the innovation, and theambition to spread its ideals as broadly as possible. Thus boundary work is very activeand diverse.

In targeting the external stakeholders, the LINSA uses a considerable amount of specialwork – organising public events, discussion forums, developing specific policy proposals,holding workshops in combination with farmers’ markets. These involve not onlyfarmers, but also activists from various civic groups, and entrepreneurs subscribing tosolidarity economy ideals. Special work can be identified also in internal boundary work,namely in training activities, in carrying our research (e.g. on price formation of organicproduce). Participants in this case are farmers, researchers, also activists. The special traitof boundary work in the Crisoperla case is the broad involvement of volunteers, and thedriving role of several key individuals, linking organizations, civic groups, and economicentities. The boundary objects demonstrate a broad range, from concepts, ideas andprinciples (of solidarity economy and organic farming, but adapted to local conditions), toevent programmes and public activities, as well as topics of learning and policy advocacy

Learning and Innovation in Agriculture and Rural Development 23

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

work, which would often turn into tangible objects (brochures, leaflets, etc.). Overall, theevolution of boundary objects can be traced from the earliest attempts to define the idealsof the LINSA as a code of practice, to very specific marketing and policy workarrangements, documents and artefacts embodying the network ideas. The outcomes ofthese activities can be termed concerted action.

The evolution of the network has been steady; successful boundary work has led toexpansion of supporters and visibility, an increased range of participants, improvedpractices and increased impact on policies promoting the dissemination of the specificinnovation (organic farming plus civic activities plus ethical entrepreneurship). In termsof network structures, this LINSA is characterized by regular and democratic decision-making (monthly meetings, supplemented by virtual communication), owing much to thedrive of some of the LINSA ‘founders’. The challenges may be seen in perfecting thegovernance of the network.

In the above two cases we have demonstrated that boundary work can foster a range ofoutcomes related to innovation dissemination and evolvement, negotiating both internaland external boundaries with diverse stakeholders. The scope and targets of boundaryactivities derive from the specific context: in the case of Care farms network boundarywork may be characterized as more specific, on a narrow range of issues, centred ontechnical knowledge. In the case of Crisoperla the range of boundary work is muchbroader and addresses social and ethical issues, aiming to link diverse actors whosubscribe to similar values, while also negotiating specific technical issues. Muchboundary work is related to network development as such.

In both cases, the boundary objects are developed in the context of specific participantsengaging in specific kinds of interaction, according to the goal of boundary work and theconnections that have to be made or managed.

4.3. Boundary Work for Sustainability

SOLINSA research suggests that negotiation of the meaning of sustainability is acomplicated endeavour, due to the diversity of LINSA members and also externalmessages from the general public and policy actors. Thus, to achieve better alignment onthe meaning of sustainability or adjust it to relevant external developments, the followingboundaries may have to be navigated: (1) between groups of LINSA stakeholders,(2) between LINSA and external civic groups, (3) between LINSA and policy actors and/or representatives of the existing regime. Boundary work on specific meanings ofsustainability and evolvement of relevant boundary objects in LINSA are presented belowin two cases.

The Brighton and Hove Food Partnership in England highlights the dynamic co-evolution and multiplicity of boundary work and objects in developing sustainable localfood systems in an urban area. Started from the need and the desire of sustainable livingin a holistic sense, the partnership aims at integrating urban food chain with sustainableactions in other related areas (waste, health, energy, transport) in order to ensure urbansustainability and wellbeing.

The Brighton and Hove City Council is one of the key driving and boundary workactors in establishing and operating the partnership. At the origin of the partnership therewas the Council’s action plan on food, which can be categorized as a specific boundaryobject and which expressed the local state’s willingness and readiness to promote

24 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

sustainable food issues. There were other boundary events and objects set up in the citybefore the establishment of the partnership: a report and a conference on food whichconcluded on the need to establish an overarching body to encourage and coordinate amore sustainable food system.

The inclusive, holistic vision of sustainable urban development formed the seminalboundary object around which a group of driving actors consisting of representativesfrom local policy and civic groups were gathered. Subsequent boundary work to developand implement this vision brought together and reconciled many individuals andnetworks with seemingly different functions and interests, but whose common valuesand beliefs were uncovered in collaborative work. Firstly boundary work happened atencounters with a broad range of civic and market actors involved. Then specific work bya task force followed – collaborative development of a proposal for the city policies andlobbying it. The boundary object evolved further to become a City Food Strategy. Afteradoption the Food Strategy still remained a boundary object as various stakeholders hadto interpret it for implementing. Satellite initiatives developed, e.g. grants to practiceurban gardening, which in themselves became boundary objects and which added anotherkind of boundary work – practice-based. Because of its holistic nature, the LINSA wasalso able to encourage other kindred networks to develop related sustainable systems(Curry and Kirwan 2014).

The holistic approach of the partnership means also that there are many diverse urbanfood actors involved in boundary interactions (there are 200 individual and collectivemembers) and many other are targeted. In order to align their attitudes and practicesactive boundary work is present. The partnership itself functions as an interface for theirinteractions and a boundary breaker. The specific boundary objects it has produced, abooklet and a food strategy, outline the diversity of boundary work and objects used inthe LINSA, which altogether are aimed at promoting sustainable living and connectingactors. They contain such general and specific items as a newsletter on food, health andenvironmental issues, a webpage, health and local food work, courses and programs topromote sustainable, healthy and active lifestyles, local produce festivals and othermeetings and networking which bring together people from diverse sectors, production ofan action plan and lobbying policies.

The boundary objects in this LINSA have evolved from the initial general vision,which later on was supplemented by more specific and practice oriented ones, allevolving over time. The range of participants is very broad and diverse, with civic groupsand local policy makers dominating. Boundary work involves co-existing encounters,specific work and practice-based connections, all giving rise to implementing a sharedtransformative idea of sustainability, of a ‘breaking the mould’ variety.

The most prominent sustainability outcome of boundary work in Brighton and Hovefood partnership is the widely shared redefined understanding of sustainable well-being inthe city and the changing food practices. The network is aiming at changing values andlifestyles, and its popularity among the citizenry shows that it is successful indisseminating its holistic vision of sustainability. The formal result of the collaborativework in the partnership has been the produced city food strategy which is adopted by thelocal state. Together with the support of the City council and the local health authoritythis adds to the legitimacy of the sustainable food system ideas and measures developedby the partnership and sends positive messages to public.

Learning and Innovation in Agriculture and Rural Development 25

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

In addition, boundary work has allowed for consolidating and strengthening the city’ssustainable food network. It has brought together a broad range of food-concerned actors,helped them to build shared values and lifestyles, reach mutual trust, agree on commoninformal rules and coordinate their actions. Regardless the common grounds andguidance that the partnership provides, its members carry out their activities in a flexiblemanner, there are no legal obligations and formal rules.

The holistic nature of the partnership poses also some barriers to reach its sustainabilitygoals. To a lesser extent this is a challenge for its internal management as members areenthusiastic and devoted but in particular it concerns establishing external links andgaining external support. For instance, it is hard to gain support from the officialagricultural knowledge system institutions and attract funding because the partnership isso multi-purpose.

The contrasting case of Biogas network in Latvia illustrates the controversies involvedin negotiating meanings of sustainability and its various dimensions, and someunintended results (or failures) of boundary work. One of the principal needs drivingboundary work in this LINSA was to arrive at societally acceptable and sustainablelocalized solutions of biogas production.

Biogas production in Latvia was politically (top-down) initiated with the arguments ofenhancing energy and agricultural sustainability. It was aimed at renewable energyproduction, relevant in the situation of increasing fossil fuel use, energy dependence,farmers’ low income and unutilized agricultural waste. However, initial boundary workand objects have been ineffective to align biogas producers with these goals and many ofthem, especially landless investors, have caught up the economic dimension of theinitiative, which was promoted by such boundary objects as production quotas, publicsubsidies and guaranteed price. In the result many of biogas plants have been developedprimarily as business projects for profit ignoring ecological and social sustainability.

The public image of biogas production has deteriorated and prioritizing businessinterests provoked protests among concerned farmers and scientists who insisted on therevision of the political, regulatory and controlling framework of biogas production toreadjust it to sustainability criteria. Sustainability of biogas production as a boundaryobject has become an explicit object of discussions among LINSA members themselves.The boundary work performed to reach sustainable local solutions of biogas productioninvolved three groups of actors: landless investors, local farmers-biogas producers, andbiogas researchers. All were interested in localized technological solutions, but not all insustainability concerns. Both farmers and researchers promote boundary work, but thegroup that stands apart is landless investors.

Boundary objects were formed of specific technologies of biogas production, eitherborrowed from abroad, or developed through own experimentation, or proposed byresearchers. Connection of actors and negotiation of technologies took several forms:encounters (between farmers and technology providers, peer to peer exchanges amongproducers), practice-based negotiations (site visits, demonstrations, training seminars),specific work (joint experimentation with researchers). The boundary objects in this caseare specific rather than general, and they constantly evolve reflecting new opportunitiesand new challenges. Recent legitimacy crisis in biogas production has prompted aninitiative, which can be labelled specific work to involve landless investors in a dialogue.However, they were not interested.

26 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

In this LINSA the boundary objects follow a trajectory from being specific and multipleto a new one, general and in acute need of reconciliation. Boundary work has beensuccessful as practice-based activities and encounters around the specific boundary objects,but to a lesser extent as specific work (a dialogue meeting, commissioned research,negotiations) on the more general idea of biogas production as a sustainable endeavour.Despite the fact that the LINSA agrees about the necessity of local sustainable solutions,there is continued disagreement on different aspects of sustainability and no commonprescriptive practices have been developed. Boundary work has not reduced thecontroversial impact of biogas production practices on environmental and social sustain-ability. Failure in better alignment of the actors can be attributed to the insufficientboundary work that would manage their considerable diversity, individualistic approachesand social distance between them and that would encourage incorporation of publicinterests of sustainability. One may also argue that the existence of powerful economic andpolitical interests at play has obstructed the adaptive course of sustainability definition.

The two cases of boundary work and objects in addressing sustainability showed theirco-development, albeit along differing trajectories. The difference in outcomes wasgreatly influenced by the inclusiveness/exclusiveness of the vision, the presence of powerimbalances, and diversity or homogeneity of actors.

5. Discussion and Conclusion

In this paper we explored the role of boundary work and boundary objects in agriculturaland rural development networks. We have presented and discussed in depth severalexamples to illustrate how actors in LINSA learn and innovate jointly for sustainability,crossing different social, institutional, and conceptual boundaries. Focusing on the role ofboundary work and objects has helped to understand how complex learning andinnovation networks function, how stakeholders negotiate sustainability, co-produceinnovation and learn for the future.

5.1. Multiple Expressions of Boundary Work Are Dynamically Emerging

The discussed LINSA demonstrate that boundary work fulfils multiple functions toachieve the network goals: as a process generating new knowledge across two or morecommunities, it strengthens the network’s internal structures, brings in new supporters,and stimulates the sharing of network ideas and the evolvement of the network’sinnovation. To foster alignment and cooperation, LINSA actors create and use boundaryobjects that represent the meaning of the negotiated idea for both themselves and theactors whose cooperation is desired. Boundary work and the evolvement of boundaryobjects are interrelated. Addressing different types of goals and actors necessitates theemergence and application of different types of boundary work and boundary objects.

SOLINSA research confirmed the existence of several forms of boundary work usedby LINSA to advance their development and socio-technical transformations. FollowingWenger’s (2000) typology, we identified encounters, boundary practices, peripheral workand specific work. We found that In LINSA these kinds of boundary work haveadditional traits, and involve specific constellations of actors, depending on the desiredoutcome and the stage in network evolution. Foremost, we note the evolvement ofboundary work types over time.

Learning and Innovation in Agriculture and Rural Development 27

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Boundary work starts as encounters and the study confirmed the great variety ofencounters: meetings, workshops, visits, seminars, etc. Encountering by doing proves tobe efficient way of boundary bridging. This often happens during field days,demonstration events, study visits and other forms of practical learning. LINSA achievemore in terms of knowledge co-creation, concerted practices and network sustainability ifpractical encounters are organized regularly. Peripheral boundary work proved relevant inhelping LINSA to build links with consumers, the like-minded civic groups and policymakers and to educate population about sustainability issues, especially if LINSA is valuebased. This kind of boundary work often takes the shape of cultural activities and publicevents, e.g. festivals, fairs, food events, science cafes, etc. Special boundary work wasobserved in many LINSA and may take a form of targeted research or demonstrationprojects, negotiations with authorities, development of a policy paper, a certification, etc.

Another aspect of boundary work is the diversity of actors; in each LINSA there is aspecific constellation of actors involved depending on the issues at hand, these involveactors from the social-economic sphere, civic, knowledge and policy spheres; no case isthe same. LINSA exhibit a particular strength in establishing links between peer-farmers,the farmers, researchers and advisors, agricultural community, civic groups and policymakers. Successful boundary work provides important gains for the network: sharedmeanings, improved practices, new supporters, visibility, and evolvement of structuresand governance. Outcomes of boundary work can be differentiated as prescriptive orflexible, where the former means that all actors undertake to implement the results ofboundary work in the same way, and latter means that actors do not commit themselves tocarrying out identical but concerted set of actions. This points to the different degrees ofcoordination of activities in networks.

5.2. Boundary Objects Take Different Shapes and Can Be Broad or Specific

The data on the boundary objects used by LINSA confirms that they are physically andconceptually varied objects embedded in the specific contexts in which the LINSAoperate. Empirically grounded types of boundary objects identified by our research are:

. Ideas, values, visions.

. Strategic papers, policy documents.

. Codes of practice, certification systems.

. Specific production methods, techniques.

. Public events (meetings, festivals, forums etc.). This kind of boundary objects isblurred with boundary interaction, but it still has its distinctive objectivation andspatial form of communicative action.

. Production and marketing infrastructure.

Within and apart of these types there is a still greater variety of boundary objects toaddress specific learning, innovation and transition to sustainability needs. Boundaryobjects may exist as shared cognitive structures (negotiated meanings), or as tangiblematerial things, objects, artefacts and reifications that embody these meanings (a leaflet ofan organization, a manual of a code of practice or a learning video for a specificproduction method), or both. The convertibility of cognitive and material sides ofboundary objects, the translation of ideas into material things consolidates the negotiatedshared knowledge, extends its durability, and fosters more effective communication of

28 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

LINSA ideas and practices internally and externally. Emerging within the LINSA andaccepted internally, boundary objects may then be introduced into interaction withexternal stakeholders to promote LINSA ideas. The range of boundary objects is broadand evolves with network development (confirming findings by Theodorakopoulos et al.2014), thus our study echoes the conclusions of Clark et al. (2011) that context is the keyfactor in boundary work success.

We also noted two broader types of boundary objects: general and specific, relatedto the stage in the development of the network, as well as various forms of boundaryobjects – ideas, documents, codes of practice, events, methods, and infrastructure. It mustbe noted that with the help of various information and communication technologies,boundary interactions, work and objects are increasingly virtual.

We have also demonstrated some network outcomes of boundary work, and theapplicability of boundary work and boundary objects concepts to analysing multi-stakeholders interactions in self-organising networks. A LINSA is a particular network, asit assembles not only farmers aiming at improving their practice (be it in terms oftechnology, marketing or organization), but many different actors, as showed in thepresented cases. In addition, a LINSA strongly focuses on the process of learning, i.e.negotiation and reflection among its members, aimed at renovation of existing practicesor introducing new solutions. Finally, learning and innovations are targeted towardsincreasing sustainability, thus the practice of a LINSA is value-laden. Accordingly,LINSA has to provide the space to negotiate these three outcomes: learning, innovation,and sustainability. We have shown that for each, boundary objects and boundary workhave particular requirements (see Figure 1).

With regard to sustainability, boundary work needs to accommodate different attitudesand practices of sustainability; with regard to learning, the boundary objects used need toenable negotiation between stakeholder groups representing various knowledge cultures(e.g. practitioners and researchers), various attitudes towards learning outcomes/mistakes,and various learning forms; with regard to innovation, boundary work is relevant tonegotiate between progressive and conservative parts of a LINSA, between LINSAmembers and potential supporters, and to adjust initial goals when they have been achievedand/or new opportunities are identified.

While most of these functions regard the internal organization of the network, there is aneed for additional boundary work between the LINSA and the outside world. Reachingthe relevant actors beyond the limits of the LINSA is crucial if innovation is aimed atinducing broader change.

Boundary work, objects and their connection are dynamic. Once proposed, negotiatedand established, boundary objects remain open for further transformations. Many caseswitness a certain sequence of boundary work and boundary objects that mirrors thedevelopment stages and related particular needs of the network. Supposedly, each of thefurther steps benefits from or is constrained by the knowledge already accumulated intothe network’s artefacts. There is certain correspondence between goals of boundary work,the kinds of boundary work employed and the boundary objects developing in theprocess. For instance, institutionalization goals often show a pattern of encounters andspecific work types of boundary work, and use highly specific boundary objects relyingon scientific knowledge. Conversely, practice-based connections tend to be forged inlearning activities, and also in dissemination of incremental innovation. This needs to beexplored in further research.

Learning and Innovation in Agriculture and Rural Development 29

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

5.3. Boundary Work Needs to Be Carefully Considered and Facilitated

Boundary interactions are not equally intensive among all network members; there arecentral and peripheral actors. Those interactions are facilitated by leaders (driving actors),individuals, or entire organizations. As LINSA include complex constellations ofstakeholders, boundary work can be potentially initiated by a range of driving actors.Often they are actors with multiple structural positions and influential personalities,whose personal and institutional traits are preconditions for engaging with LINSA in jointlearning and innovation for sustainability in a range of roles – networker, facilitator,participatory researcher, boundary person, or specialist expert. These actors may betermed transition partners – to emphasize the multiplicity of their context-specific role(Moschitz et al. 2014). As boundary work develops from one kind to another, its drivingactors may change as well.

Our study suggests that boundary practices not necessarily stem from formalassignments of individuals and organizations, for example – corporative responsibilitiesof advisory service. These roles are rather attributed, negotiated and agreed amongnetwork partners and are based on their merit, trustful relations and committedengagement in practical interactions. Therefore boundary persons are often trustedresearchers, civic activists and farmers who are dedicated to LINSA objectives andinspire other participants to share knowledge.

The multiple requirements that boundary objects need to fulfil for LINSA to functionefficiently call for a careful consideration and facilitation of boundary work. As the

Sustainability• different a�tudes• different prac�ces

Innova�on• progressive and

conserva�ve members

• members and possible supporters

• ini�al goals and new innova�on goals

Learning• different knowledge

cultures (e.g. research - prac�ce)

• different a�tudes towards learning

• different learning forms

Figure 1. Requirements of boundary objects in LINSA: aspects boundary work has to bridge.

30 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

results indicate, boundary objects can emerge from interaction, and, in line with earlierfindings by Klerkx et al. (2012), they can be created purposefully. In both cases, strategicfacilitation of boundary work will help an effective functioning of a LINSA. Schneideret al. (2009) emphasize the need to respect the structures in which the different actorsusually work in order to achieve a successful engagement in boundary work acrossorganizations. To support boundary work in LINSA it is thus necessary to carefullyconsider the backgrounds of different members. Boundary work will need the moreattentive facilitation the more heterogeneous the members are, also to make effective useof heterogeneity.

However, one also needs to realize the limitations of boundary objects: they do notnecessarily connect all the different life worlds represented in LINSA, there can beexcluded actors or backsliders (resembling findings by Kimble et al. 2010; Klerkx et al.2012). This is partly related to the next limitation of boundary objects, i.e., in many casesthey are issue specific and not relevant to every member, contested and variouslyprioritized depending on needs and priorities. Not all actors equally participate in theirconstruction (typically LINSA leaders and strategic partners are more active). Boundaryobjects can be poorly distributed or inaccessible to all members. They cease to exist orturn into another boundary objects when the original learning need is satisfied. Boundaryobjects are not a universal remedy for LINSA problems but pragmatic entities to be co-constructed in order to solve learning and innovation issues.

From facilitation and innovation support perspective, boundary work can be used asan instrument for LINSA transition towards sustainability, and it can be doneaccording to two different logics. One logic is to concentrate on a specific boundaryobjects in LINSA, develop them through specifically targeted boundary interactions(e.g. development projects) driven by active boundary spanning individuals or organiza-tions. Such an intervention would lead to fast learning and dissemination of novelpractices, and stimulate radical innovations through establishing prescriptive codes andmethods. However, the impact of such boundary work may be limited to narrowgroups of stakeholders. The other logic is based on broader range of driving actors,uses of shared network resources, and boundary work that is embedded in differentnetwork practices and gradually develops and refines many different boundary objectsresponding to network diversity and scope of innovation. This kind of boundary workmay produce a slower transition towards sustainability, yet the outcomes may bemore durable.

Our examples have shown that boundary work supports also reflection in LINSA, asmeanings of learning, innovation and sustainability are continuously negotiated andaccepted as its goals. Through continued boundary work, reflection becomes a definingfeature for LINSA. Facilitation is again needed to help this reflection while producinginnovation. Researchers can take the role of facilitators who organize boundary work,meeting spaces and even suggest boundary objects. Yet, our work with LINSA showedthat researchers need to be careful not to conflate roles. Within LINSA they can becomepart of the network, thus engage themselves in boundary work beyond mere facilitation ofthis process, as described also by Huzzard, Ahlberg, and Ekmann (2012). Reflectivity ofLINSA then involves reflective researchers who facilitate boundary work of LINSAmembers to co-produce outcomes in learning, innovation and sustainability.

Learning and Innovation in Agriculture and Rural Development 31

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Acknowledgements

We acknowledge the contribution of SOLINSA colleagues whose research reports and LINSAdescriptions have been used in the preparation of this article.

Funding

This work was supported by the European Commission under the 7th Research FrameworkProgramme [grant number 266306].

References

Bruckmeier, K., and H. Tovey. 2008. “Knowledge in Sustainable Rural Development: From Formsof Knowledge to Knowledge Processes.” Sociologia Ruralis 48 (3): 313–329. doi:10.1111/j.1467-9523.2008.00466.x.

Brunori, G., D. Barjolle, A-C. Dockes, S. Helmle, J. Ingram, L. Klerkx, H. Moshitz, G. Nemes, andT. Tisenkopfs. 2013. “CAP Reform and Innovation: The Role of Learning and InnovationNetworks.” Eurochoices 12 (2): 27–33. doi:10.1111/1746-692X.12025.

Carolan, M. S. 2006. “Social Change and the Adoption and Adaptation of Knowledge Claims:Whose Truth Do You Trust in Regard to Sustainable Agriculture?” Agriculture and HumanValues 23 (3): 325–339.

Carr, A., and R. Wilkinson. 2005. “Beyond Participation: Boundary Organizations as a New Spacefor Farmers and Scientists to Interact.” Society and Natural Resources 18 (3): 255–265.doi:10.1080/08941920590908123.

Cash, D. W., W. C. Clark, F. Alcock, N. M. Dickson, N. Eckley, D. H. Guston, J. Jager, and R. B.Mitchell. 2003. “Knowledge Systems for Sustainable Development.” Proceedings of theNational Academy of Sciences of the USA. Accessed December 12. http://www.pnas.org/content/100/14/8086.full.

Clark, W. C., T. P. Tomich, M. van Noordwijk, D. Guston, D. Catacutan, N. M. Dickson, andElizabeth McNie. 2011. “Boundary Work for Sustainable Development: Natural ResourceManagement at the Consultative Group on International Agricultural Research (CGIAR).”Proceedings of the National Academy of Sciences of the USA. Accessed June 14. http://www.pnas.org/content/early/2011/08/11/0900231108.full.pdf+html.

Curry, N., and J. Kirwan. 2014. “The Role of Tacit Knowledge in Developing Networks forSustainable Agriculture.” Sociologia Ruralis 54 (3): 341–361. doi:10.1111/soru.12048.

Gieryn, T. F. 1983. “Boundary-work and the Demarcation of Science from Non-science: Strains andInterests in Professional Ideologies of Scientists.” American Sociological Review 48 (6): 781–795.

Guston, D. H. 2001. “Boundary Organizations in Environmental Policy and Science.” ScienceTechnology and Human Values 26 (4): 399–408. doi:10.1177/016224390102600401.

Hermans, F., I. Horlings, P. J. Beers, and H. Mommaas. 2010. “The Contested Redefinition of aSustainable Countryside: Revisiting Frouws’ Rurality Discourses.” Sociologia Ruralis 50:46–63. doi:10.1111/j.1467-9523.2009.00501.x.

Hornidge, A-K., M. Ul-Hassan, and P. P. Mollinga. 2009. “Follow the Innovation’: A JointExperimentation and Learning Approach to Transdisciplinary Innovation Research.” ZEFWorkingPaper Series 39. Accessed December 4. https://www.econstor.eu/dspace/bitstream/10419/88376/1/605235813.pdf.

Huzzard, T., B. M. Ahlberg, and M. Ekmann. 2012. “Constructing Interorganizational Collaboration:The Action Researcher as Boundary Subject.” Action Research 8 (3): 293–314. doi:10.1177/1476750309335206.

Kimble, C., C. Grenier, and K. Goglio-Primard. 2010. “Innovation and Knowledge Sharing acrossProfessional Boundaries: Political Interplay between Boundary Objects and Brokers.”International Journal of Information Management 30 (5): 437–444. doi:10.1016/j.ijinfomgt.2010.02.002.

Klerkx, L., and C. Leeuwis. 2009. “Ehe Emergence and Embedding of Innovation Brokers atDifferent Innovation System Levels: Insights from the Dutch Agricultural Sector.” Technolo-gical Forecasting and Social Change 76: 849–860. doi:10.1016/j.techfore.2008.10.001.

32 T. Tisenkopfs et al.

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15

Klerkx, L., S. van Bommel, B. BO, H. Holster, J. V. Zwartkruis, and N. Aarts. 2012. “DesignProcess Outputs as Boundary Objects in Agricultural Innovation Projects: Functions andLimitations.” Agricultural Systems 113: 39–49. doi:10.1016/j.agsy.2012.07.006.

Klerkx, L., B. van Mierlo, and C. Leeuwis. 2012. “Evolution of Systems Approaches toAgricultural Innovation: Concepts, Analysis and Interventions.” In Farming Systems Researchinto the 21 st Century: The New Dynamic, edited by I. Darnhofer, D. Gibbon, and B. Dedieu,457–483. Dordrecht: Springer.

Knickel, K., G. Brunori, S. Rand, and J. Proost. 2009. “Towards a Better Conceptual Frameworkfor Innovation Processes in Agriculture and Rural Development: From Linear Models toSystemic Approaches.” Journal of Agricultural Education and Extension 15 (2): 131–146.doi:10.1080/13892240902909064.

Koutsouris, A. 2008. “Innovating Towards Sustainable Agriculture: AGreek Case Study.” The Journalof Agricultural Education and Extension 14: 203–215. doi:10.1080/13892240802207619.

Lave, J., and E. Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. New York:Cambridge University Press.

Leeuwis, C., and N. Aarts. 2011. “Rethinking Communication in Innovation Processes: CreatingSpace for Change in Complex Systems.” The Journal of Agricultural Education and Extension17: 21–36. doi:10.1080/1389224X.2011.536344.

Mollinga, P. P. 2010. Boundary Work and the Complexity of Natural Resources Management.Accessed December 4. https://dl.sciencesocieties.org/publications/cs/pdfs/50/Supplement_1/S-1.

Moore, K. 1996. “Organizing Integrity: American Science and the Creation of Public InterestOrganizations, 1955–1975.” American Journal of Sociology 101 (6): 1592–1562. doi:10.1086/230868.

Moschitz, H., T. Tisenkopfs, G. Brunori, R. Home, I. Kunda, and S. Sumane. 2014. Final Report onthe SOLINSA Project. SOLINSA Deliverable N°8.2 of the SOLINSA project, GA Nr. 266306.

Oreszczyn, S., A. Lane, and S. Carr. 2010. “The Role of Networks of Practice and Webs ofInfluencers on Farmers’ Engagement with and Learning about Agricultural Innovations.”Journal of Rural Studies 26: 404–417. doi:10.1016/j.jrurstud.2010.03.003.

Renting, H., H. Oostindie, C. Laurent, G. Brunori, D. Barjolle, and A. M. Jervell. 2008.“Multifunctionality of Agricultural Activities, Changing Rural Identities and New InstitutionalArrangements.” International Journal of Agricultural Resources, Governance and Ecology 7(4–5): 361–385. doi:10.1504/IJARGE.2008.020083.

Ring, J. K., A. M. Peredo, and J. J. Chrisman. 2010. “Business Networks and EconomicDevelopment in Rural Communities in the United States.” Entrepreneurship Theory andPractice 34 (1): 171–195. doi:10.1111/j.1540-6520.2009.00307.x.

Santos, B., J. A. Nunes, and M. P. Meneses. 2007. “Opening up the Canon of Knowledge andRecognition of Difference.” In Another Knowledge is Possible: Beyond Northern Epistemol-ogies, edited by B. S. Santos. London: Verso. Accessed February 12. http://www.boaventuradesousasantos.pt/media/Introduction(3).pdf.

Schneider, F., P. Fry, T. Ledermann, and S. Rist. 2009. “Social Learning Processes in Swiss SoilProtection—the ‘From Farmer - To Farmer’ Project.” Human Ecology 37 (4): 475–489.doi:10.1007/s10745-009-9262-1.

Star, S. L., and J. R. Griesemer. 1989. “Institutional Ecology, ‘Translations’ and Boundary Objects:Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology.” Social Studies ofScience 19 (3): 387–420. doi:10.1177/030631289019003001.

Theodorakopoulos, N., D. Bennett, and D. J. Sánchez Preciado. 2014. “Intermediation forTechnology Diffusion and User Innovation in a Developing Rural Economy: A SocialLearning Perspective.” Entrepreneurship & Regional Development 26 (7–8): 645–662.doi:10.1080/08985626.2014.971077.

Tisenkopfs, T., I. Kunda, and S. Šūmane. 2014. “Learning as Issue Framing in AgriculturalInnovation Networks.” The Journal of Agricultural Education and Extension 20 (3): 1–18.doi:10.1080/1389224X.2014.887759.

Wenger, E. 2000. “Communities of Practice and Social Learning Systems.” Organization 7 (2):225–246. doi:10.1177/135050840072002.

Learning and Innovation in Agriculture and Rural Development 33

Dow

nloa

ded

by [

Wag

enin

gen

UR

Lib

rary

] at

12:

15 1

2 Ja

nuar

y 20

15